Abstract
The air-sea drag coefficient controls the transfer of momentum from wind to water. In modeling storm surge, this coefficient is a crucial parameter for estimating the surge height. This study uses two strong wind events on Lake Erie to calibrate the drag coefficient using the Coupled Ocean Atmosphere Wave Sediment Transport (COAWST) modeling system and the the Regional Ocean Modeling System (ROMS). Simulated waves are generated on the lake with Simulating WAves Nearshore (SWAN). Wind setdown provides the opportunity to eliminate wave setup as a contributing factor, since waves are minimal at the upwind shore. The study finds that model results significantly underestimate wind setdown and storm surge when a typical open-ocean formulation without waves is used for the drag coefficient. The contribution of waves to wind setdown and storm surge is 34.7%. Scattered lake ice also increases the effective drag coefficient by a factor of 1.1.
Highlights
Strong winds acting on a body of water for a sustained number of hours may produce currents and a shift in the entire water mass
These two phenomena are opposite in vertical direction and comparable in magnitude; wind setdown is a drop in the water level, and storm surge is a rise in water level
The RUC 252 data product used here must be interpolated from 20 km to match the 808 m average grid cell size in the Lake Erie domain, and that interpolation cannot capture the finer wind features that occur on length scales smaller than 20 km
Summary
Strong winds acting on a body of water for a sustained number of hours may produce currents and a shift in the entire water mass. Wind setdown occurs when the body of water recedes from the upwind shoreline; storm surge occurs when the water shifts toward the downwind shoreline and inundates dry land. Heaps published a study of a twodimensional model of storm surge on the North Sea.[1] He used a rectangular model grid oriented on latitude and longitude lines, with the open-sea and coastal boundaries of northwestern Europe. His grid resolution was about 34 km per grid cell, and his calculating time step was 6 and 3 minutes. Wind forcing (speed and direction) at his grid points produced changes in the free surface zeta and in the u- and v-components of the mean current
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
